IL NUOVO CIMENTO
DOI 10.1393/ncc/i2013-11407-8
Vol. 36 C, N. 1
Gennaio-Febbraio 2013
Colloquia: IFAE 2012
Particle discrimination in TeO2 bolometers
N. Casali(1 )(8 ), J. W. Beeman(2 ), F. Bellini(3 )(4 ), L. Cardani(3 )(4 ),
I. Dafinei(4 ), S. Di Domizio(5 )(6 ), F. Ferroni(3 )(4 ), F. Orio(4 ), G. Pessina(7 ),
G. Piperno(3 )(4 ), S. Pirro(7 )(8 ), C. Tomei(4 ) and M. Vignati(4 )
(1 )
(2 )
(3 )
(4 )
(5 )
(6 )
(7 )
(8 )
Università dell’Aquila - L’Aquila, Italy
Lawrence Berkeley National Laboratory - Berkeley, CA 94720, USA
Dipartimento di Fisica, Sapienza Università di Roma - Roma I-00185, Italy
INFN, Sezione di Roma - Roma I-00185, Italy
Dipartimento di Fisica, Università di Genova - Genova I-16146, Italy
INFN, Sezione di Genova - Genova I-16146, Italy
INFN, Sezione di Milano-Bicocca - Milano I-20126, Italy
INFN, Laboratori Nazionali del Gran Sasso - Assergi, L’Aquila I-67010, Italy
ricevuto il 31 Agosto 2012
Summary. — TeO2 crystals are used as bolometric detectors in experiments searching for the neutrinoless double beta decay. The extreme rarity of the studied signal
requires an ultra low background level. The natural radioactivity of the materials represents for these experiments the main background source, in particular the
component due to α particles. In this work, we present a possibility for particle
discrimination in TeO2 bolometers using the Čerenkov radiation. The results show
that we are able to measure the light produced by β/γ particles, which can be explained as due to Čerenkov emission. No light is detected from α particles, allowing
the rejection of this background source.
PACS 14.60.Pq – Neutrino mass and mixing.
PACS 07.57.Kp – Bolometers; infrared, submillimeter wave, microwave, and
radiowave receivers and detectors.
PACS 23.40.-s – β decay; double β decay; electron and muon capture.
PACS 29.40.Ka – Cherenkov detectors.
1. – Research for neutrinoless double beta decay with TeO2 bolometers
The experiments that want to understand if the neutrino is a Majorana particle search
for a nuclear reaction known as neutrinoless double beta decay (0νββ) [1]. In the 0νββ
a nucleus decays into another nucleus emitting two electrons and no anti-neutrinos, thus
violating the total lepton number. The decay is mediated by the exchange between two
β vertexes of a neutrino, which controls the decay rate through its mass (mββ ). The
c Società Italiana di Fisica
!
57
58
N. CASALI, J. W. BEEMAN, F. BELLINI, ETC.
signature of this process is a monochromatic line at the Q-value of the decay in the sum
energy spectrum of the two electrons.
The sensitivity of an experiment to the 0νββ half-life goes as
(1)
S ∝ a#
!
Mt
,
B∆E
where a is the isotopic abundance, # the efficiency of the signal, M the detector mass, t
the observation time, B the background level and ∆E the energy resolution.
It is clear from eq. (1) that the experimental investigation of this process requires a
large amount of 0νββ emitter and ultra-low background detectors. The bolometric technique can satisfy these requirements, as has been demonstrated by the CUORICINO [2]
detector in such a convincing way that a 1-ton scale experiment, CUORE [3], is now
under construction. A bolometer is a low-temperature–operated calorimeter able to fully
thermalize the energy released by a particle, providing excellent energy resolution and
low-energy thresholds.
The CUORE experiment will use 988 TeO2 bolometers of 750 g each to study the
0νββ decay of 130 Te (34.2% isotopic abundance) featuring an energy resolution ∆E of
5 keV FWHM at the Q-value of the decay (2527.518 ± 0.013 keV [4]).
Although TeO2 crystals are extremely good bolometers, they are unable to operate
any discrimination between α particles and β/γ ones [5]; unfortunately, the radioactive
contaminants, located on the surfaces of the detector or on passive elements facing them,
produce α particles that can lose part of their energy in a few microns and deposit the
rest in a bolometer. This leads to an essentially flat background that affects also the
Region of Interest (ROI) around the 0νββ Q-value [6].
The advantages offered by this material in terms of bolometric performances and
natural isotopic abundance are a strong motivation to pursue an extremely challenging
option: the read-out of the Čerenkov light emitted in β/γ interactions (and not in α
ones) to perform a particles discrimination, as has been pointed out in ref. [7].
2. – Search for Čerenkov light from TeO2 bolometer
We operated as bolometer a 3.0×2.4×2.8 cm3 TeO2 crystal, whose weight was 116.65 g.
The crystal was doped with natural samarium, which contains 15% of 147 Sm, a long living
isotope (T1/2 = 1.07 × 1011 y) that undergoes α decay with a Q-value of 2310 ± 1 keV.
This decay allows a direct analysis of the behavior of α’s in an energy region close to
the 0νββ [8]. We faced to the TeO2 crystal a high sensitivity dark bolometer acting
as a light detector (LD) [9]. This consisted of a 66 mm diameter 1 mm thick pure Ge
crystal, covered with a 600 Å layer of SiO2 to ensure good light absorption. A reflecting
foil (3M VM2002) was placed around the TeO2 crystal to enhance the light collection.
The detector was operated deep underground in the Gran Sasso National Laboratories
(Italy), in the CUORE R&D cryostat. The details of the experimental setup and of the
data analysis technique can be found in [10, 11].
The detector was exposed to a 232 Th source placed outside the cryostat, to provide
statistics to the β/γ sample, and to perform the energy calibration of the TeO2 bolometer.
We used the calibration data to estimate the light emitted from γ interactions, and
the background data to estimate the light from α interactions. The distributions of the
light measured with the LD (EL ) for the two peaks close to the 0νββ energy, the γ-ray
at 2615 keV and the α decay at 2310 keV are shown in fig. 1 left. The distribution of EL
59
2310 keV α (Sm-147)
Entries
599
χ2 / ndf
10.906 / 11
A
113.997 ± 5.976
mean
0.004 ± 0.004
σ
0.103 ± 0.004
100
80
60
EL [keV]
120
0.5
0.4
0.3
0.2
0.1
0
-0.1
-0.2
-0.3
EL [keV]
counts / 0.05 keV
PARTICLE DISCRIMINATION IN TeO2 BOLOMETERS
0.5
0.4
0.3
0.2
0.1
0
-0.1
-0.2
-0.3
40
20
40
35
30
25
20
0
0.1
0.2
0.3
0.4
0.5 0.6
EL2 [keV]
2615 keV γ (Tl-208)
Entries
179
χ2 / ndf
6.276 / 8
A
37.407 ± 3.482
mean
0.195 ± 0.007
σ
0.093 ± 0.005
15
10
5
0
-0.4 -0.3 -0.2 -0.1
0
232
Th calibration
<EL> [keV]
counts / 0.05 keV
0
-0.4 -0.3 -0.2 -0.1
background
0.1
0.2
0.3
0.4
0.5 0.6
EL2 [keV]
0.20
0.15
0.10
0.05
0.00
-0.050
1000
2000
3000
4000
5000
6000
Energy[keV]
Fig. 1. – Left: Distribution of the light emitted from 2310 keV α decay (top) and 2615 keV γ-ray
interactions (bottom), fitted with a Gaussian function. Right: Distribution of the emitted light
EL versus particle energy. In the top (background data) and middle (calibration data) figures
all the peaks are marked in black. In the bottom figure the average light for each peak is shown,
using background data for the α peaks (triangles) and calibration data for the γ peaks (circles);
γ and α points are fitted with two different lines.
versus particle energy is shown in fig. 1 right. We estimated the average light of each
peak, #EL $ and we performed separate linear fits for α’s and β/γ’s to the distribution of
#EL $ versus energy (bottom of the figure).
The light measured for α particles is close to zero, which is expected if the light we
detected from γ interactions is due to Čerenkov radiation; for β particles with energy
close to the 0νββ Q-value we estimate an average light energy #EL0ν $ = 173 eV.
3. – Conclusions
In this work we showed that a background suppression can be obtained detecting
the Čerenkov light. By applying a threshold T on the detected light (EL > T ), we can
decrease the α background and increase the sensitivity of the experiment. The sensitivity
increase is proportional to the separation between signal and α particles, defined as the
distance of #EL0ν $ from zero energy in units of the energy resolution σ. The separation
obtained in this work is ∼ 2σ, while one needs at least 5. A further improvement can
be obtained by reducing the noise, which seems to be dominated by the read-out of
the bolometer (thermistor, cables or electronics) or by increasing the efficiency of light
collection which, in this work, was estimated to be only ∼ 20%.
REFERENCES
[1] Avignone F. T., Elliott S. R. and Engel J., Rev. Mod. Phys., 80 (2008) 481.
[2] Andreotti E. et al., Astropart. Phys., 34 (2011) 822.
[3] Ardito R. et al., arXiv:hep-ex/0501010 (2005).
60
N. CASALI, J. W. BEEMAN, F. BELLINI, ETC.
[4] Redshaw M., Mount B. J., Myers E. G. and Avignone F. T., Phys. Rev. Lett., 102
(2009) 12502.
[5] Arnaboldi C. et al., Astropart. Phys., 34 (2011) 797.
[6] Pavan M. et al., Eur. Phys. J. A, 36 (2008) 159.
[7] Tabarelli de Fatis T., Eur. Phys. J. C, 65 (2010) 359.
[8] Bellini F. et al., JINST, 5 (2010) P12005.
[9] Pirro S. et al., Nucl. Instrum. Methods A, 559 (2006) 361.
[10] Beeman J. W. et al., Astropart. Phys., 35 (2012) 558.
[11] Piperno G., Pirro S. and Vignati M., JINST, 6 (2011) P10005.